- From: Robin Berjon <robin@berjon.com>
- Date: Thu, 16 Feb 2012 15:57:06 +0100
- To: James Graham <jgraham@opera.com>
- Cc: Arthur Barstow <art.barstow@nokia.com>, public-webapps-testsuite@w3.org, 김성택 <stkim@infraware.co.kr>, ext 유성옥 <sungok.you@infraware.co.kr>, 이종흔 <jhlee@infraware.co.kr>
On Feb 16, 2012, at 15:51 , James Graham wrote: > On Thu 16 Feb 2012 03:46:12 PM CET, Arthur Barstow wrote: >> As such, I think a consequence of this requirement is that reference to the testharness resources must use an absolute path and that means using one of: >> >> 1. <script src="/resources/testharness.js"> >> 2. <script src="http://w3c-test.org/resources/testharness.js "> >> >> Would our active testing folks - e.g. Kris, James, Ms2ger, Aryeh, PLH - please confirm this requirement and the path that should be used? > > I think 1) is preferred style here. There isn't really a style that never causes problems but 2) is particularly problematic because it means everyone needs to edit the files to reuse them outside the W3C infrastructure (a strong requirement for us). Relative paths can also be problematic because it encourages copying testharness.js around rather than having a single maintained copy. Agreed. An additional issue with relative paths is that you have to change them when you move tests from /spec/submitted/someone to /spec/approved. Not a huge change but exactly the type that will waste one's time. (1) isn't perfect in that you need to run a local server to run tests locally, but I think that's pretty much a requirement if you want to be realistic anyway. -- Robin Berjon - http://berjon.com/ - @robinberjon Coming up soon: I'm teaching a W3C online course on Mobile Web Apps http://www.w3devcampus.com/writing-great-web-applications-for-mobile/
Received on Thursday, 16 February 2012 14:57:39 UTC